Patent Application
20230394198
Aspects of the disclosure relate to digital twin modeling. In some cases, models, such as machine learning models, may be used for prediction. Due to limited computing resources, processing power, and/or machine learning model capabilities, the number of features considered by such models may be limited. For example, engineers, administrators, and/or other individuals may perform feature engineering to identify a subset of features, related to a use case that may have the greatest impact on the model's output. In such instances, however, remaining features may be disregarded. This may result in identification of an output that might not be as accurate as a hypothetical output in which all features were considered by the model. Accordingly, it may be advantageous to train, generate, and/or otherwise host a model capable of analyzing all related features of a use case (e.g., without limiting the number of features through feature engineering), while balancing the limitations of computing resources such as available memory, processing power, and/or other resources.
Aspects of the disclosure provide effective, efficient, scalable, and convenient technical solutions that address and overcome the technical problems associated with modeling physical property and forecasting value. In accordance with one or more embodiments of the disclosure, a computing platform comprising at least one processor, a communication interface, and memory storing computer-readable instructions may receive historical property information. The computing platform may train, using the historical property information, a digital twin property evaluation engine, configured to model a physical property based on characteristics of the physical property using a computer simulation, where: 1) training the digital twin property evaluation engine further configures the digital twin property evaluation engine to output event processing information for the physical property, and 2) training the digital twin property evaluation engine comprises generating a knowledge graph, where each node of the knowledge graph may be a machine learning model and each edge of the knowledge graph may represent a relationship between features corresponding to each machine learning model. The computing platform may receive, from a client device, an event processing request identifying a first physical property. The computing platform may generate, using the digital twin property evaluation engine, a computer simulation of the first physical property. The computing platform may execute, over period of time for simulation, the computer simulation of the first physical property to output event processing information for the first physical property. The computing platform may send, to the client device, the event processing information and one or more commands directing the client device to display the event processing information, which may cause the client device to display the event processing information.
In one or more instances, the historical property information may include one or more of: a number of bedrooms, a number of bathrooms, location details, weather impacts, floorplans, effects of daily use, a zip code, maintenance costs, dates of repairs, and/or other information. In one or more instances, the machine learning models may include one or more of: a climate change model, a credit history model, a stock market model, and/or other models.
In one or more examples, the machine learning models may be characterized by different time scales. In one or more examples, the machine learning models may output information for each of a plurality of features, feature engineering may be needed to provide a model output by a single machine learning model, and/or the feature engineering may cause at least one of the plurality of features to not be analyzed by the single machine learning model.
In one or more instances, the relationships between the features may indicate how each feature affects other features. In one or more instances, the digital twin property evaluation engine may automatically learn the relationships over time.
In one or more examples, the relationships may be manually defined. In one or more examples, the event processing information may indicate one or more of: loan information or a risk score indicating a level of risk associated with providing a loan for the first physical property.
In one or more instances, the event processing information may include an explanation of the loan information or the risk score. In one or more instances, the event processing information may indicate a second physical property, with a predetermined number of matching features to the first physical property, along with loan information for both the first physical property and the second physical property, and the second physical property may have a lower risk score than the first physical property, and a lower interest rate than the first physical property.
These features, along with many others, are discussed in greater detail below.
The present disclosure is illustrated by way of example and not limited in the accompanying figures in which like reference numerals indicate similar elements and in which:
In the following description of various illustrative embodiments, reference is made to the accompanying drawings, which form a part hereof, and in which is shown, by way of illustration, various embodiments in which aspects of the disclosure may be practiced. In some instances, other embodiments may be utilized, and structural and functional modifications may be made, without departing from the scope of the present disclosure.
It is noted that various connections between elements are discussed in the following description. It is noted that these connections are general and, unless specified otherwise, may be direct or indirect, wired or wireless, and that the specification is not intended to be limiting in this respect.
As a brief introduction to the concepts described further herein, one or more aspects of the disclosure describe a digital twin based home evaluation engine to help customers make loan decisions. Home mortgage may be one of the most important financial decisions a bank customer may make in their lifetime. The decision may involve uncertainties on both the lender and borrower sides, as many events may occur during the duration of a loan (which may, e.g., be 15-years).
Owning a home may be expensive, and may involve repaying a loan consisting of principle and interest, as well as paying local taxes, insurances, and/or other costs. In addition, there may be costs for heating or cooling the home and other utilities, as well as regular maintenance costs.
A lender may usually make loan assumptions based on the current situations and circumstances, which may lead to a conservative decision, denying loans to people who may be qualified, and thus losing the potential income from such loans. Sometimes, the lender may, in spite of their cautions, misjudge the paying capacity of the borrowers. Although the borrow may have the income sufficient to pay the mortgage and taxes, the cost of maintaining a home may involve utilities (e.g., heating, cooling, and the like) maintenance, repair, and/or additional burdens on the borrower. As a consequence, the home might not be properly repaired, and may increase a chance of a bad or otherwise defaulted loan. The borrower, on the other hand, may be optimistic about their ability to repay the loan.
Accordingly, it may be advantageous to develop a tool that may help users make a reasonable estimate of the costs of home ownership to help facilitate loan decisions. A digital twin of a home model may help both the lender as well as the borrower to make a reasonable estimate of the costs of ownership of a home to facilitate loan decisions.
A digital twin model may be designed to accurately reflect a physical object. The object being studied (e.g., a house or other property) may be outfitted with various sensors related to vital areas of functionality. These sensors may produce data about different aspects of the physical object's performance, such as energy output, weather conditions, and/or other aspects. This data may then be related to a processing system and applied to the digital copy.
Once informed with such data, the virtual model may be used to run simulations, study performance issues, generate possible improvements, and/or perform other actions, all with the goal of generating valuable insights, which may then be applied back to the original physical object.
Although simulations and digital twins both utilize digital models to replicate a system's various processes, a digital twin may actually be a virtual environment, which may make it considerably richer for study. The difference between digital twin and simulation may be a matter of scale. For example, while a simulation may study a single process, a digital twin may itself run any number of useful simulations in order to study multiple processes.
Furthermore, simulations might not benefit from having real-time data. Digital twins, however, may be designed around a two-way flow of information that may first occur when object sensors provide relevant data to the system processor, and then insights created by the processor may be shared back with the original source object.
The system described herein supports a plug and play digital twin of a home that may simulate costs of maintenance and/or other repair and outcome based simulations, predictions and/or estimates to help make loan decisions at different times and/or for different loan durations.
The digital twin of the home may be a plug and play mode that may be extended for home features such as number of bedrooms, bathrooms, den, living area, basement, heated and cooled areas, and/or other information. In some instances, the digital twin may also consider other external factors that may affect home prices such as weather, location, climate change, and/or other information.
In some instances, the model may accept internet of things sensor data if available from a home. If the data is not available, it may be extrapolated and/or otherwise customized for any type of home. In the latter case, it may use publicly available heating and/or cooling data to build an intelligent prediction model.
In some instances, the user may be able to provide the plan of a home that may be digitally read and used to make a prediction.
The plan may be generated during home appraisal, or may be available from a public record of previous home sales or deeds. If such information is not available, information from similar homes in the area may be used to make a prediction.
The user may then provide the duration for which the simulation is needed. It may, for example, make cost estimates for winter (e.g., November to April) after five years, so that the user may obtain an estimate of an older boiler heating a 3000 square foot house.
Accordingly, described herein is a data and knowledge driven digital twin based simulation engine to help customers and lenders make lending decisions based on a customizable simulation of houses or other real property based on different scenarios. The system may improve valuation of debt to income and/or risk estimation for significant expenditures based on lifespan of everything from the roof to the water heater. The exact home information for each home might not be needed. Rather, a digital twin may be created to simulate all types of homes. This plug and play digital twin simulation model may be used for home appraisal and loan configuration/formulation. The model may be extended for size of the home (e.g., number of bedrooms, bathrooms, etc.), year built, location, climate and weather, school district ratings, and/or other information. The model may be used to calculate long term maintenance costs (e.g., heating, air conditioning, repair costs, and/or other costs), and may identify whether the applicant is able to afford the cost of the loan in addition to the maintenance or other ancillary costs.
As described further below, digital twin host platform 102 may be a computer system that includes one or more computing devices (e.g., servers, server blades, or the like) and/or other computer components (e.g., processors, memories, communication interfaces) that may be used to train, host, and/or otherwise refine a digital twin model, which may, e.g., include a knowledge graph linking together a plurality of individual feature models. In these instances, nodes of the knowledge graph may represent the feature models, and edges between the nodes may represent relationships between the feature models (e.g., how the operation and/or outputs of each model affects the others).
Information source system 103 may include one or more computing devices and/or other computer components (e.g., processors, memories, communication interfaces). In some instances, information source system 103 may include one or more data sources that may store historical information having an effect on property valuations and/or risk (e.g., property features, maintenance costs, weather/climate information, location information, stock market information, previous sale information, appliance information, deterioration information, cost information, school district information, and/or other information). Additionally or alternatively, the information source system 103 may include one or more real time sensors (e.g., internet of things sensors and/or other sensors), which may, e.g., provide real time property information. In some instances, these sensors may be located in or around various properties.
Client device 104 may be a mobile device, tablet, smartphone, desktop computer, laptop computer, and/or other device that may be used by an individual (such as a potential client of a financial institution, who may, e.g., be searching for a property loan) to request a loan and/or provide information. In some instances, the client device 104 may be configured to provide one or more user interfaces (e.g., loan interfaces, or the like).
Enterprise user device 105 may be a mobile device, tablet, smartphone, desktop computer, laptop computer, and/or other device that may be used by an individual (such as an underwriter, loan officer, and/or other employee financial institution, who may, e.g., be responsible for approving or denying loan requests) to process loan requests. In some instances, the enterprise user device 105 may be configured to provide one or more user interfaces (e.g., loan interfaces, or the like).
Computing environment 100 also may include one or more networks, which may interconnect digital twin host platform 102, information source system 103, client device 104, enterprise user device 105, or the like. For example, computing environment 100 may include a network 101 (which may interconnect, e.g., digital twin host platform 102, information source system 103, client device 104, enterprise user device 105, or the like).
In one or more arrangements, digital twin host platform 102, information source system 103, client device 104, and/or enterprise user device 105 may be any type of computing device capable of sending and/or receiving requests and processing the requests accordingly. For example, digital twin host platform 102, information source system 103, client device 104, enterprise user device 105, and/or the other systems included in computing environment 100 may, in some instances, be and/or include server computers, desktop computers, laptop computers, tablet computers, smart phones, or the like that may include one or more processors, memories, communication interfaces, storage devices, and/or other components. As noted above, and as illustrated in greater detail below, any and/or all of digital twin host platform 102, information source system 103, client device 104, and/or enterprise user device 105 may, in some instances, be special-purpose computing devices configured to perform specific functions.
Referring to
Digital twin module 112a may have instructions that direct and/or cause data digital twin host platform 102 to execute advanced techniques to generate digital twin simulations of physical properties and to run the simulations so as to identify costs, risks, and/or other information about the properties. In some instances, the digital twin module 112a may include a knowledge graph supported by (and storing relationships between) one or more feature models (which may, in some instances, be machine learning and/or other models). Digital twin database 112b may store information used by digital twin module 112a and/or digital twin host platform 102 in application of advanced techniques to generate and/or otherwise simulate the digital twins of physical property described above, and/or in performing other functions.
At step 202, the information source system 103 may send historical information to the digital twin host platform 102. For example, the information source system 103 may send the historical information to the digital twin host platform 102 while the first wireless data connection is established. For example, the information source system 103 may send home layout information (e.g., number of bedrooms, number of bathrooms, floorplans, and/or other information), historical repair information (e.g., date of last repair of appliances/structures, and/or other information), cost information (e.g., initial costs, maintenance costs, insurance costs, repair costs, and/or other information), location information (e.g., zip code, location specific average values, and/or other information), deterioration information (e.g., effects of daily use, effects of weather, and/or other information), market information (e.g., stock information, interest rate information, property market trends, and/or other information), school district information, and/or other information that may be used to identify property values and/or costs (and/or changes to these values/costs).
At step 203, the digital twin host platform 102 may receive the historical information sent at step 202. For example, the digital twin host platform 102 may receive the historical information via the communication interface 113 and while the first wireless data connection is established.
Although steps 202 and 203 are described with regard to historical information, the information source system 103 may, in some instances, additionally or alternatively send real time information (similar to the historical information) at any point throughout the illustrative event sequence without departing from the scope of the disclosure. For example, real time information may be sent from one or more data sources, which may, e.g., include internet of things sensors, and/or other sensors/data sources.
At step 204, the digital twin host platform 102 may train or otherwise configure a digital twin generation engine to generate a digital twin (e.g., a computer simulation) of physical property. For example, an enterprise organization such as a financial institution may support customer purchases of physical property by granting a loan, which the customer may pay back to the financial institution over the life of the loan, which may, e.g., be 15 years, 30 years, or another period of time. Over the life of the loan, a value of the physical property may change (e.g., due to deterioration, everyday use, market trends, and/or other information). Accordingly, in determining whether or not to issue a loan, and/or an associated interest rate, it may be important for the financial institution to consider a risk of loss not only at the current time, but over the life of the loan. Furthermore, because financial institutions may make such decisions for a large number of properties, it may be advantageous to avoid the need to visit each property individually, and to instead simulate or otherwise model the properties. Thus, it may be advantageous to train a model configured to execute a digital simulation of the physical property over a period of time corresponding to the loan, and to output information based on the simulation.
In some instances, in configuring the digital twin generation engine, the digital twin host platform 102 may configure a plug and play model, which may allow the input of various parameters to create a simulated version of various properties based on characteristics and/or other information of the property (which may, e.g., be included with an event processing request or otherwise). In doing so, the digital twin host platform 102 might not create a unique digital twin generation engine for each property, but rather may be able to simulate any property type based on various received inputs. This may achieve certain technical advantages such as conserving memory (e.g., when compared to a database storing digital twins for each property individually).
In some instances, in training the digital twin generation engine, the digital twin host platform 102 may further configure the digital twin generation engine to output a risk score, loan options, similar properties, recommendations, and/or other information based on simulation of a given digital twin, once generated. For example, the digital twin host platform 102 may configure the digital twin generation engine to simulate a property over time (e.g., 30 years), and may output the risk score, loan options, similar properties, recommendations, and/or other information based on results of the simulation.
To do so, the digital twin host platform 102 may generate a knowledge graph that includes the various features for consideration in determining value of a property over time. In some instances, these features may correspond to the various types of historical information described above. In these instances, the digital twin host platform 102 may train models (e.g., machine learning models and/or other models) that may be used to output information corresponding to a particular feature over a particular time period. For example, the digital twin host platform 102 may train, using the historical information, a weather model, configured to analyze weather patterns, precipitation information, wind speeds, sun exposure, and/or other weather information. As another example, the digital twin host platform 102 may train, using the historical information, a pricing model, configured to analyze average property values, market trends, property layout information, and/or other information. Additionally or alternatively, the digital twin host platform 102 may train, using the historical information, a deterioration model, configured to analyze effects of daily use, repair dates, lifespan of appliances/structures/other features, and/or other information. Additionally or alternatively, the digital twin host platform 102 may train, using the historical information, a predicted costs model, configured to analyze historic trends, income ratios, maintenance costs, variable rates, insurance costs, and/or other information. Additionally or alternatively, the digital twin host platform 102 may train, using the historical information, an interest rate model, configured to analyze market trends, historical rates, consumer information, and/or other information to produce a recommended interest rate. Additionally or alternatively, the digital twin host platform 102 may train other models configured to provide outputs that may inform loan decisions (e.g., a credit history model, stock market model, and/or other models). In these instances, the digital twin host platform 102 may represent such trained models as nodes of the knowledge graph. In some instances, these feature models may be trained using one or more supervised and/or unsupervised learning techniques.
By training each model separately, the digital twin host platform 102 may consider information corresponding to each model in producing outputs (e.g., rather than limiting the amount of information considered using feature engineering, as is described further below). Furthermore, each model may have a different time scale, and thus the features corresponding to each model may have their own unique time scales accordingly.
Similarly, the digital twin host platform 102 may use edges of the knowledge graph (e.g., connecting the nodes) to represent relationships between the various models. For example, the edge between the weather model and the pricing model may indicate that higher average precipitation amounts cause lower property values, or that a particular orientation of the windows results in greater interior sun exposure. As another example, the edge between the deterioration model and the pricing model may indicate that higher deterioration causes lower property values. Similarly, the edge between the deterioration model and the predicted costs model may indicate that higher deterioration likelihood causes an increase in predicted costs (which may thus, in turn, result in lower property values). The knowledge graph may include a number of such edges between all of the models (not merely limited to the illustrative examples described above), which may indicate the interplay between the results of each model.
In some instances, these rules and/or relationships of the knowledge graph edges may be manually defined. Additionally or alternatively, the digital twin host platform 102 may automatically learn these relationships by analyzing the results of the various models for various properties, so as to identify the most powerful correlating factors over time. For example, the digital twin host platform 102 may automatically set a rule if a predetermined number of digital twin simulations results in the given correlation (e.g., precipitation values that exceed a given threshold result in a lower price when compared to similarly situated properties with precipitation values that do not exceed the given threshold).
In some instances, one of the feature models within the knowledge graph may be a risk model, configured to output a risk score indicating a level of risk associated with a loan on the particular property. In these instances, there may be a relationship between the level of risk of the property and one or more other features, as represented by an edge between the node of the risk model and the nodes of the remaining models. For example, as price of a property decreases over time, the corresponding risk score may increase, and vice versa.
Accordingly, the digital twin host platform 102 may train the digital twin generation engine to output a risk score for a given property based on its simulation (e.g., if price drops a first predetermined amount over 30 years, a first risk score may be output, if price drops a second predetermined amount, greater than the first predetermined amount, over 30 years, a second risk score, higher than the first risk score, may be output, etc.). Similarly, the digital twin host platform 102 may train the digital twin generation engine to output interest rates and/or other loan information based on the identified risk score. For example, the digital twin host platform 102 may train the digital twin generation engine to output a first interest rate if the risk score is greater than or equal to a first risk threshold, a second interest rate if the risk score is greater than or equal to a second risk threshold (less than the first risk threshold), but less than the first risk threshold, and so on.
At step 205, the client device 104 may establish a connection with the digital twin host platform 102. For example, the client device 104 may establish a second wireless data connection with the digital twin host platform 102 to link the client device 104 to the digital twin host platform 102 (e.g., in preparation for sending an event processing request). In some instances, the client device 104 may identify whether or not a connection is established with the digital twin host platform 102. If a connection is already established with the digital twin host platform 102, the client device 104 might not re-establish the connection. If a connection is not yet established with the digital twin host platform 102, the client device 104 may establish the second wireless data connection as described herein.
Referring to
At step 207, the digital twin host platform 102 may receive the event processing request sent at step 206. For example, the digital twin host platform 102 may receive the event processing request via the communication interface 113 and while the second wireless data connection is established.
Although the transmission and receipt of the event processing request and other information are described at steps 206 and 207, additional information such as weather information, usage information, and/or other information may also be sent to and received by the digital twin host platform 102 without departing from the scope of the disclosure. In some instances, information may be sent by the client device 104, information source system 103, and/or one or more sensors (e.g., IoT sensors, or the like).
At step 208, the digital twin host platform 102 may identify a risk score and/or other information for the property. For example, the digital twin host platform 102 may identify the features for which information is received (e.g., usage information, weather information, cost information, floorplan information, location information, and/or other information), and may feed information for the given features into their corresponding models within the digital twin generation engine. In doing so, the digital twin host platform 102 may generate a digital twin simulation for the property, which may, e.g., be able to model impact/effects on the property over time. Once generated, the digital twin host platform 102 may execute the digital twin simulation over a period of time, which may, e.g., correspond to a life of a loan requested in the event processing request. In executing the digital twin simulation, the digital twin host platform 102 may identify the resulting effects on a value of the home due to deterioration, usage, inclement weather, market trends, and/or other information (e.g., by executing the various feature models and utilizing the relationships between such models embedded in the edges of the knowledge graph).
Once a value has been predicted, the digital twin host platform 102 may compare a difference value, indicating a difference between the current price of the property and a final price of the property (e.g., at the end of the loan term), to one or more predetermined thresholds to identify a risk score. For example, if the price increased, the digital twin host platform 102 may assign a risk score of 0. If the price decreased less than 10%, the digital twin host platform 102 may assign a risk score of 3. If the price decreased between 10-30%, the digital twin host platform 102 may assign a risk score of 6. If the price decreased more than 30%, the digital twin host platform may assign a risk score of 9.
Once a risk score has been output, the digital twin host platform 102 may identify an interest rate for the requested loan based on the risk score (e.g., by comparing the risk score to one or more predetermined risk thresholds). For example, if the risk score is less than 1, the digital twin host platform 102 may assign a best or lowest available interest rate (e.g., an interest rate of 2.8%). If the risk score is less than 5, but greater than 1, the digital twin host platform 102 may assign next best or next lowest interest rate (e.g., an interest rate of 3.5%). If the risk score is greater than 5, the digital twin host platform 102 may assign a next best or next lowest interest rate (e.g., an interest rate of 5%).
Additionally or alternatively, the digital twin host platform 102 may identify, using the various feature models, one or more similarly situated properties (e.g., has a matching number of bedrooms, bathrooms, and/or other floorplan information, within a common price range (e.g., $5000 range, or the like), and/or is located within the same zip code). In these instances, the digital twin host platform 102 may repeat the above described simulation analysis to output potential risk scores/interest rates for these identified properties. For example, the digital twin host platform 102 may identify a similarly situated property with less risk due to less anticipated deterioration. In these instances, the digital twin host platform 102 may identify various property options for a consumer, which may, e.g., have lower interest rates.
By performing this analysis, the digital twin host platform 102 may efficiently and accurately generate loan decisions. In addition to reducing time spent to output loan information (e.g., an automated process may be quicker than manual analysis), the results may be more accurate (e.g., the digital twin generation engine may be capable of considering more features, variables, and/or other information than a human actor). Additionally, using the digital twin generation engine may provide additional technical advantages over other automated techniques such as pure machine learning. For example, in machine learning, due to limitations of processors, models, and/or other computing resources, the number of features considered in a machine learning model may be limited through feature engineering (e.g., so as to select features with the greatest impact on the model output, while causing at least one other feature not to be analyzed by the machine learning model). In some instances, however, although the remaining features may have less impact on the machine learning output than the selected features, consideration of such remaining features may nevertheless increase accuracy of the final output. Accordingly, by training different models for each feature individually (e.g., rather than a single model trained to consider all features), all such features may be considered in the digital twin analysis. Furthermore, rather than merely outputting individual results from each model, by connecting the feature models through edges of a knowledge graph, the relationships between the outputs of each feature model may also be taken into account in the analysis. As a result, a property may be simulated over time while taking into account all features that may have an impact on risk of loss, interest rates, and/or other information for the property, as well as the relationships between such features, thus increasing accuracy over traditional machine learning methods as well as balancing the limits of memory, processors, and/or other computing resources.
At step 209, the digital twin host platform 102 may establish a connection with the enterprise user device 105. For example, the digital twin host platform 102 may establish a third wireless data connection with the enterprise user device 105 to link the digital twin host platform 102 with the enterprise user device 105 (e.g., for purposes of sending notifications and/or responses to the event processing request). In some instances, the digital twin host platform 102 may identify whether a connection is already established with the enterprise user device 105. If a connection is already established with the enterprise user device 105, the digital twin host platform 102 might not re-establish the connection. If a connection is not yet established with the enterprise user device 105, the digital twin host platform 102 may establish the third wireless data connection as described herein.
At step 210, the digital twin host platform 102 may send a notification to the client device 104 and/or the enterprise user device 105 that includes the output of the digital twin analysis (e.g., event processing information). For example, the digital twin host platform 102 may send a notification that includes the rates and/or other properties identified to the client device 104 and/or enterprise user device 105 (which may e.g., allow a customer and/or a loan officer/underwriter to view the information). In some instances, the digital twin host platform 102 may also send information explaining the rates. For example, the digital twin host platform 102 may indicate a feature of the property that primarily contributed to a determination of high risk (such as high maintenance costs, likelihood of deterioration, market trends, or the like). In some instances, the digital twin host platform 102 may send the notification to the client device 104 and/or the enterprise user device 105 via the communication interface 113 and while the second and/or third wireless data connections are established. In some instances, the digital twin host platform 102 may also send one or more commands directing the client device 104 and/or the enterprise user device 105 to display the notification.
At step 211, the client device 104 and/or the enterprise user device 105 may receive the notifications sent at step 210. For example, the client device 104 and/or the enterprise user device 105 may receive the notifications while the second and/or third wireless data connections are established. In some instances, the client device 104 and/or the enterprise user device 105 may also receive the one or more commands directing the client device 104 and/or the enterprise user device 105 to display the notifications.
Referring to
At step 213, the digital twin host platform 102 may update the digital twin generation engine based on the outputs generated at step 208 and/or any feedback received from the client device 104 and/or enterprise user device 105 (e.g., received based on the notifications displayed at step 212). In doing so, the digital twin host platform 102 may dynamically and continuously update and/or otherwise refine the digital twin generation engine so as to increase accuracy of the feature models, knowledge graph, simulations, and/or other aspects of the digital twin generation engine.
Although the above described method is primarily described in the context of an initial home loan, the method may be applied to commercial properties, refinance situations, leases, and/or other situations related to property loans without departing from the scope of the disclosure.
One or more aspects of the disclosure may be embodied in computer-usable data or computer-executable instructions, such as in one or more program modules, executed by one or more computers or other devices to perform the operations described herein. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types when executed by one or more processors in a computer or other data processing device. The computer-executable instructions may be stored as computer-readable instructions on a computer-readable medium such as a hard disk, optical disk, removable storage media, solid-state memory, RAM, and the like. The functionality of the program modules may be combined or distributed as desired in various embodiments. In addition, the functionality may be embodied in whole or in part in firmware or hardware equivalents, such as integrated circuits, application-specific integrated circuits (ASICs), field programmable gate arrays (FPGA), and the like. Particular data structures may be used to more effectively implement one or more aspects of the disclosure, and such data structures are contemplated to be within the scope of computer executable instructions and computer-usable data described herein.
Various aspects described herein may be embodied as a method, an apparatus, or as one or more computer-readable media storing computer-executable instructions. Accordingly, those aspects may take the form of an entirely hardware embodiment, an entirely software embodiment, an entirely firmware embodiment, or an embodiment combining software, hardware, and firmware aspects in any combination. In addition, various signals representing data or events as described herein may be transferred between a source and a destination in the form of light or electromagnetic waves traveling through signal-conducting media such as metal wires, optical fibers, or wireless transmission media (e.g., air or space). In general, the one or more computer-readable media may be and/or include one or more non-transitory computer-readable media.
As described herein, the various methods and acts may be operative across one or more computing servers and one or more networks. The functionality may be distributed in any manner, or may be located in a single computing device (e.g., a server, a client computer, and the like). For example, in alternative embodiments, one or more of the computing platforms discussed above may be combined into a single computing platform, and the various functions of each computing platform may be performed by the single computing platform. In such arrangements, any and/or all of the above-discussed communications between computing platforms may correspond to data being accessed, moved, modified, updated, and/or otherwise used by the single computing platform. Additionally or alternatively, one or more of the computing platforms discussed above may be implemented in one or more virtual machines that are provided by one or more physical computing devices. In such arrangements, the various functions of each computing platform may be performed by the one or more virtual machines, and any and/or all of the above-discussed communications between computing platforms may correspond to data being accessed, moved, modified, updated, and/or otherwise used by the one or more virtual machines.
Aspects of the disclosure have been described in terms of illustrative embodiments thereof. Numerous other embodiments, modifications, and variations within the scope and spirit of the appended claims will occur to persons of ordinary skill in the art from a review of this disclosure. For example, one or more of the steps depicted in the illustrative figures may be performed in other than the recited order, and one or more depicted steps may be optional in accordance with aspects of the disclosure.
